Only accepted work is recordedAboutGuidesContactSign in
CitePepCitePep
The Inner Workings of Machine Learning Explained
Correction Explainers

The Inner Workings of Machine Learning Explained

Corrected by Emir Baycan · on When Notes Fly · 11 July 2026 · View published page ↗

How does machine learning work: AlexNet's 15.3 percent error against 26.2 for second place started the deep learning era.

Misleading wording

Changes accepted

1 edited

Computed from the content the publisher accepted, line by line. Neither the contributor nor CitePep writes this diff by hand.

edit
−had error rates for dark-skinned women that were up to 34.7 percent higher than for light-skinned men
+had error rates for dark-skinned women as high as 34.7 percent, versus under 1 percent for light-skinned men

Why this is better

Verified live article (post 1292, how-does-machine-learning-work): body already correctly reads 'as high as 34.7 percent, versus under 1 percent for light-skinned men'. Checked FAQ (4-step op-sequence, 5 items) and excerpt independently - neither mentions this Gender Shades statistic. Already correct, no PUT needed.

More by Emir Baycan in Explainers

All of Emir Baycan's contributions →